Verified65% confidenceFactExact time
The size of the KV cache is determined by three multiplied dimensions: the size of each record, how many tokens are combined into one record, and whether layers can share caches with each other.
3
Sources
65%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWith KV Cache, each round of Q projection only needs to compute a 1×D_k vector, whereas without caching, a full N×D_k matrix must be computed.75% similarUnverifiedKV缓存的规律是：上下文长度翻倍缓存大致翻倍，并发满上下文请求翻倍缓存再翻倍74% similarVerifiedKV缓存将已计算的Key-Value矩阵存储在显存中实现增量计算，代价是显存占用随序列长度线性增长74% similarVerifiedThe VRAM consumption of the KV cache is proportional to context length, number of model layers, and number of attention heads; under an 8K context, a typical 27B model may consume an additional 2–4 GB of VRAM.74% similarUnverifiedPagedAttention divides the KV Cache into fixed-size blocks, allocates them dynamically on demand, and allows blocks from different requests to be stored non-contiguously.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/900717API
curl https://kongchang.com/api/v1/knowledge/claims/900717MCP
get_claim(id=900717)